The present invention relates to a picture synthesizing system and, more particularly, to a picture synthesizing system for synthesizing main picture data compressed by the MPEG (Moving Picture coding Experts Group, ISO/IEC) scheme with sub-picture data such as subtitles compressed by the run-length scheme.
Recently, a system for decoding and synthesizing main picture data compressed by the MPEG scheme and sub-picture data such as subtitles on a movie or karaoke video, and playing back the resultant data has been proposed. In such a system, when MPEG decoding, run-length decoding, and synthesis processing are to be processed by software, a sub-picture data conversion section performs format conversion of sub-picture data, but color palette conversion is not performed.
In this case, however, in synthesis processing, YUV data (Y: luminance signal; U and V: color difference signals) must be rearranged in accordance with a display format by looking up a table. This increases the load on the CPU (Central Processing Unit).
It is an object of the present invention to provide a picture synthesizing system which attains a reduction in processing load by eliminating the necessity to rearrange YUV data in synthesis processing for main picture data and sub-picture data.
In order to achieve the above object, according to the present invention, there is provided a picture synthesizing system comprising first decoding means for decoding main picture data by a first picture compression scheme, second decoding means for outputting a color table and frame data by decoding sub-picture data by a second picture compression scheme, first conversion means for converting the color table from the second decoding means on the basis of a display format in which a synthesized picture is displayed, second conversion means for converting the frame data from the second decoding means, and picture synthesizing means for generating sub-picture data first on the basis of the color table and frame data output from the first and second conversion means, and then synthesizing the sub-picture data with the main picture data output from said first decoding means.